A Spam Email Detection Mechanism for English Language Text Emails Using Deep Learning Approach
Source of Publication
Proceedings of the Workshop on Enabling Technologies: Infrastructure for Collaborative Enterprises, WETICE
© 2020 IEEE. Phishing emails are emails that pretend to be from a trusted company that target users to provide personal or financial information. Sometimes, they include links that may download malicious software on user's computers, when clicked. Such emails are easily detected by spam filters that classify any email with a link as a phishing email. However, emails that have no links, link-less emails, requires more effort from the spam filters. Although many researches have been done on this topic, spam filters are still classifying some benign emails as phishing and vice-versa. This paper is focused on classifying link-less emails using machine learning approach, deep neural networks. Deep neural networks differs from simple neural network by having multiple hidden layers where data must be processed before reaching the output layer. The data used in this research is publicly available online. Hyper parameter optimization, was performed, using different settings on the data. In order to demonstrate the effectiveness of the approach, precision, recall and accuracy were computed. The results show that the deep neural network performed well in many of its settings.
deep learning, email spammers, machine learning, spam detection, text-based analysis
Kaddoura, Sanaa; Alfandi, Omar; and Dahmani, Nadia, "A Spam Email Detection Mechanism for English Language Text Emails Using Deep Learning Approach" (2020). All Works. 4076.
Indexed in Scopus